Carnegie Mellon
What isa Data Center?
As an organization’s IT needs to grow, the total cost of
ownership (TCO) of it’s IT resources grows
Resource management, power, cooling scaling, fault‐
tolerance, space (real estate) becomes an issue
Solution? Consolidate your IT resources into a data
center!
Introduction to Cloud Computing
5.
Carnegie Mellon
Data CenterGoals
Introduction to Cloud Computing
Minimize floor space, maximize resource density
Keep the equipment in a safe and controlled environment
Resources must be easy to access physically and manage,
components must be individually replaceable without
disrupting normal operations
Optimize operational costs
6.
Carnegie Mellon
Physical Layout
Equipment is placed in racks
Equipment is designed in a modular fashion to fit into
rack units (1U, 2U etc)
Servers, blade enclosures, networking and power
equipment etc can be loaded into these racks
1U Server
Introduction to Cloud Computing
7U Blade center
7.
Carnegie Mellon
Data Centercomponents
Introduction to Cloud Computing
Air conditioning
▪ Keep all components in the manufacturer’s recommended
temperature range
Redundant Power
▪ UPS/Generators
▪ Multiple power feeds
Fire protection
Physical security
▪ CCTV/Access Control
Connectivity
▪ Multiple ISPs/Leased Lines
Monitoring Systems
Carnegie Mellon
Servers
Computersthat provide “services” to “clients”.
Typically more powerful and have more resources than
desktop computers.
Organizations requiring more and more physical servers
to provide various services (Web, Email, Database, etc.)
Server hardware is becoming more compact.
Introduction to Cloud Computing
10.
Carnegie Mellon
Servers forDatacenters
For large‐scale installations, server hardware is getting
more compact to help with manageability, scalability, and
power and cooling
Organizations would like to conserve the amount of floor
space dedicated to their data centers.
Introduction to Cloud Computing
11.
Carnegie Mellon
Blades andBlade Enclosures
A blade server is a stripped down server computer with a
modular design
A blade enclosure holds multiple blade servers and
provides power, interfaces and cooling for the individual
blade servers.
A single rack can hold up to 42 1U servers
Introduction to Cloud Computing
Carnegie Mellon
Our CloudHardware
⚫ IBM Bladecenter H
▪ Advanced Management Module
▪ Two Nortel Gigabit Switches
▪ 112 cores
▪ 7TB of storage
⚫ 14 Blades, each having:
⚫
⚫
⚫ Two quad‐core Intel Xeon(R) CPU (8 Cores)
8GB RAM
Two 300GB SAS disks
Introduction to Cloud Computing
14.
Carnegie Mellon
……….
13 VMhosts Blade #14 (Control)
Hardware Overview
Qloud Blades
▪ Network: 2 (1Gbit switches)
▪ Each blade is connected to each of the switches
1 Gbit switch
1 Gbit switch
Introduction to Cloud Computing
15.
Carnegie Mellon
Hardware Overview
Each Blade
8 GB RAM
speed is 667MHz
300 GB SAS
SCS 1 disk
300 GB SAS
SCS 1 disk
CPU
Intel Xe on(R)
3.16GHz & 64-bit
Bandwidth: 21.3GB/s
Bandwidth: 600MB/s
CPU
Intel Xe on(R)
3.16GHz & 64-bit
Introduction to Cloud Computing
16.
Carnegie Mellon
Cloud SoftwareStack
Introduction to Cloud Computing
⚫ IBM Cloud Software V1.4
⚫ Tivoli Provisioning Manager (TPM) V5.1.1
⚫ Manages software installation of VMs
⚫ System provisioning
⚫ Cloud UI V1.4‐GA
⚫ Web interface to TPM
⚫ IBM Directory Server V6.0
⚫ User management and LDAP
⚫ DB2 V8.1
⚫ Storage for TPM, Cloud UI and Directory Services
⚫ RedHat Enterprise Linux (RHEL) 5.1
⚫ Primary OS
⚫ Xen 3.0.3
⚫ VM Hypervisior
17.
Carnegie Mellon
NFS
(VM)
* LDAP
*DB2 V8.1
* IBM Directory
Server V6.0
(VM)
* TPM V5.1.1
* DB2 V8.1
*Cloud UI V1.4-GA
(VM)
Gateway
(VM)
Xen 3.0.3
RedHat Enterprise Linux (RHEL) 5.1
Blade
Introduction to Cloud Computing
Software Stack Overview
The Control Blade
18.
Carnegie Mellon
Endpoint SoftwareStack
⚫ RedHat Enterprise Linux 5.2
⚫ OS
⚫ Hadoop 0.16.4 / 0.20.1
⚫ Distributed Environment
⚫ IBM Java SDK 1.6.0
⚫ Programming API
⚫ Cloud Monitoring Agent V6.1.0
⚫ To report status of VMs to TPM
⚫ WebSphere Application Server v6.1, Linux
⚫ For developing SaaS applications
⚫ DB2 Server V8.1 For Linux
⚫ Database server for SaaS applications
Not necessary in this Course
Introduction to Cloud Computing
19.
Carnegie Mellon
Software StackOverview
The 13 VM hosts
Hadoop 0.16.4 / 0.20.1
WebSphere Application Server v6.1 for Linux
DB2 Server V8.1 For Linux VM
IBM Java SDK 1.6.0
Cloud Monitoring Agent V6.1.0
RedHat Enterprise Linux 5.2
Xen 3.0.3
RedHat Enterprise Linux (RHEL) 5.1
Blade
Introduction to Cloud Computing